coded phosphotransferase signal was 
also found in the blood. Therefore, the 
neo* gene is active in the circulating 
hematopoietic system of at least some of 
these animals four months after bone 
marrow infection and transplantation. 
It is not apparent why we have ob- 
tained so much higher efficiency (even in 
the absence of IL-3) of infection and in 
vivo expression compared to the work of 
others. One explanation may be stability 
differences between N2 and the retro- 
viral vectors used by others. A second 
possibility is that, although little strain 
difference was found here between the 
DBA/2J and NFS/N mouse lines, the 
efficiency of infection of the bone mar- 
row of other mouse strains (for instance. 
C3H/HeJ 4 ) is substantially different. Per- 
haps the 12- to 14-day foci examined by 
Williams tl al. (4) have a lower infection 
frequency than the 10-day foci we stud- 
ied. A final potential difference is in the 
reported liters of the vectors. If actual 
titers do differ significantly between lab- 
oratories {J). then marked apparent dif- 
ferences in bone marrow infection effi- 
ciency r >ild be produced. 
To be applicable for gene therapy, 
vectors such as N2 would carry addition- 
al gene sequences. Such additional se- 
quences may have potential detrimental 
effects on tiler. The results here indicate, 
however, that titers as low as 6 * Id 4 
cfu/ml still infect murine hematopoietic 
stem cells with high enough efficiency to 
have possible therapeutic value. 
In a complementary study by Keller ri 
al. HO) it was demonstrated that the N2 
vector can be found integrated and ex- 
pressed in all the blood cell lineages in 
long-term reconstituted mice, including 
T and B lymphocytes. Similar studies 
with a slightly different vector have re- 
cently been published by Dick mil. ill). 
In a number of earlier studies it was 
shown that retroviral vectors can under- 
go rearrangements and/or deletions. Us- 
ing a vector derived from Friend spleen 
focus-forming vims. Joyner and Bern- 
stein reported deletions of either (he 
inserted thymidine kinase sequences or 
of the viral env gene {12). Others have 
reported similar problems of rearrange- 
ment in their vectors (4. 13). Although 
we have not yet sequenced the provims. 
analyses such as those presented here, 
with Southern blots. Tl ribonuclease and 
neo* gene product assays suggest an 
intact vector structure both during initial 
infection of stem cells and during subse- 
quent in vivo stem cell proliferation and 
differentiation. 
The utility of retroviruses as vectors 
for the high efficiency transfer of exoge- 
nous genetic sequences into hematopoi- 
mwk 
ctic cells has potential elinieal relevance 
(/). A number of genetic diseases are 
known where the primary effect is upon 
the hematopoietic system. This report 
has established the conditions for the 
high efficiency transfer and expression of 
a gene into murine bone marrow using a 
new retroviral vector. 
krfimm and San 
I W. F. Anderson. Warr 224. 401 (19*4); A. 
Bemoan. S Berger. 0 Huvur. J. Dick, m 
(Irarta tar larrnag Bnmipln ma4 MriAndi. 
J. Set low end A. Hoilacndcr. Eds. iPIcnum. 
New Yurt. I9*<|. vsil 7. p 235. 
2. A. Joyner. G Keller. II. A. Phillips. A. Berrv- 
stan. Saiarr Ibtajaai JgS. 5SA ( 19*31. 
J. A D Miller. R J. Eckncr. 0. J. Jolly. T. 
Fncdman. I. M. Vemu. Sciratr 22S. AM I I9*4|. 
4 D A W.llums. |. K Lemivchka. 0 C Nalhan. 
R. C. Mulligan, Saiarr Ihmjaal Jit. 476 
119*41 
5. E. Gdh»a. M Kolhe. K Noonnn, R Kucherla- 
pati. J Viral. 44. *4 s i| 9*2>. L.-H. S. Hwu» 
end E. Cislhua. M St. 417 1 19*41; L.-H. 5. 
Hwang. J Part. E. (iilhon. Hal. CrU BaU. 4. 
121N 1 1 VH4 ) 
6. r' Mann. R l'. Mulligan. D Baltimore. CrU JJ. 
IJJtlWUl. 
7. J. N. I Me if al..J ImmamJ IJI. 212 (1993): J. 
N I hie. J Keller, t. Henderson. F. Klein. E. 
Palavrynski. JN4 129. 24311 19«2l 
t. G Cathala rr al . USA 2. J2* I ISRJi. 
». K. 7jnn. I). DiMumi. T. Maniaiis. CrU M. HAS 
I 19*3). 
10 G. Keller. C. Paige. E. Gilhoa. E. S. Wagner. 
Naiarr Ihmjimt. in press. 
It. J. E. Dick. M C. Magli. D Husjar. R. A. 
Phillips. A. Bern slan. CrU 42. 71 I 19*5). 
12. A. L Joyner end A. Bernstein. Hal. CrU Biui. 
3. 2191 M9*3). 
IJ. C. L. Ccpko. 8. F.. Roberts. R. C. Mulligan. 
CrU 31. lost ||9*4|: M Emerman end H M. 
Temia. J. Viral. Jt. 42 1 19*41. 
14 p. J. Southern and P. Berg. .Wider. Appl. (Iran. 
I. J27CIM2I. 
15. M. Wigler. A. PeDicer. S. Silvcrstcin. R. A»el. 
CrU 14. 72J ( I979|. 
16. M. Gross- Bellard. P (Met. P. Chamhon. far. 
J. Batchrm. J6. 32 I I97J|. 
17. R. K. Humphries e; ml.. Am. J. Ham. Crart.JI. 
295 (19*5). 
II. E. Southern. J. Hal. Biol. fg. 503 ( |97J|. 
19. B Reiss. R. Sprengel. H. WUI. H. Schaller. 
Crmr 31. 211 I IW4). 
20. Wc thank Jamc\ IMc for purifed IL-3. helpful 
discussions, and for critically reading the manu- 
script. Sandra Ruscctti and Linda Wolff for on- 
going discussions: are also thank Judy Di Pietro 
and Shen Bemoan for cicctlcnt technical as- 
sistance. M. Joan Curcio for help with the Tl 
nhonuc lease analysis. Jeanne McLachlin and 
James Zwiebel for assistance with the DNA 
analyses. David Trauher for help in establishing 
the bone marrow transfer techniques. Sheldon 
Goldberg for assistance in the early pan of the 
wort, and Gordon Keller for advice. Supported 
■a pan lE.G.) by grants from the Ame::._a 
Cancer Society iMV-lIbB) and the March of 
Dunes (l-RSAl. 
27 June I9R5: accepted 14 October 1913 
SCIENCE. VOL 2.VI 
Recombinant DNA Research, Volume 12 
[239] 
